dp-optimizer
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted algorithmic code and transitions, creating a potential surface for indirect prompt injection.
- Ingestion points: External data enters the agent context through the
dpCode,stateDefinition, andtransitionsfields in the input schema defined inSKILL.md. - Boundary markers: The instructions do not specify delimiters or prompts to ignore embedded instructions within the ingested code.
- Capability inventory: The skill has access to file system tools including
Read,Write,Grep,Glob, andEdit. - Sanitization: There is no evidence of input sanitization or validation to filter malicious content from the provided code strings.
Audit Metadata